Building Bhakashal - Magic Distribution

Image by Judson Huss

I have been asked why Bhakashal is an AD&D retroclone, and it is largely due to the fact there are a few systems in the game that I find elegant and well designed and worth building off of. 

The treasure system in AD&D is one of those things.

I have a particular interpretation of it, and it’s best explained in a blog post.

So here’s the short version, Gygax created random encounter tables, weighted tables, Monster/Treasure type tables and Treasure tables. If you use those to randomly generate all of your treasure, in conjunction with a modified version of the “Generating a party at the spur of the moment” rules for NPC treasure, you will get a particular distribution of treasure in your game. 

If you read what Gygax has to say about it, he suggests it needs to be heavily curated, and here I disagree. I think his design, which has several layers of randomness (random treasure odds by monster, random determination of treasure type, randomness on each individual treasure type table), makes magic items less common than they are in most campaigns, and when they do occur, are more often temporary or consumable magic items and small bonus weapons. 

I have run this system for a while, some of my groups have been at it for three years, some for two or one, but every player has had 80 hours with this system at the table per game year. I use GP=XP, XP for magic items, variable XP progress by class and low XP for monsters. The average level’s of the PCs in campaigns are: 1 year of gaming - 4-5th lv, 2 years - 5-6th, 3 years 6-8th, with a year being 80 hours, 40 two hour games. I have tested different things at different times, but I have used the treasure system, and the random generation of spells system, since day 1.

I consider this to be a representative sample of results from running the magic system as I do, with over 40 players in 15 separate campaigns over a three year period.

The System

When the players meet an NPC I have outfitted them according to a modified version of the rules for creating NPCs on the Spur of the Moment from Appendix P of the 1e DMG. I leave all the odds alone, and just change out the fixed lists based on the campaign items. This system filters out the more powerful magic items based on level, so a high level NPC might have a powerful item or two.

Here are a few of the tables:

When players have a planned encounter with an intelligent monster, I generate the treasure according to treasure type, and the monster will sometimes use the items if appropriate, anything not appropriate will be at their lair. I sometimes roll for this rather than just automatically check (perhaps the intelligent monster doesn’t know that the sword is magical, not all magic powers in items in AD&D manifest to the eye or the touch). This can produce more powerful items for the intelligent monsters to use. 

When using random encounters, which are rolled for whenever the party is traveling from point A to point B, or in a dangerous place like a dungeon, for intelligent creatures there will be pre-rolled items because they should have a chance to use them, but for other sorts of monsters, I roll treasure on the spot. 

So for example, an encounter with an owlbear or a chimera will not involve rolling up treasure ahead of time to see if they will be using it, for those I roll at the table in game. This treasure will be “in lair”. Sometimes the encounter is not in lair (I roll for this using the odds in the monster manual entries), if so, unless they have a tracker to find the lair, they don’t get a crack at treasure at all. 

Tracking to the lair involves possible encounters along the way, but assuming that they find it, I roll to see if there are any other inhabitants, also, any wounded monsters that fled might have come back to their lair to heal and rest. I roll for that.

Assuming they get to the treasure, I roll randomly for it, by treasure type (which is determined by Monster type). Note that all animals, tigers, bears, wolves, jaguars, etc. all dinosaurs, many low HD monsters like giant ants, skeletons and zombies, and even a few high HD monsters like Aerial Servants , have no treasure at all. So some random encounters are going to present no loot. If you look at the random encounter tables you will notice that there are a lot of animals/dinosaurs and low HD monsters with no loot at all. 

The odds of random encounters producing treasure are thus further reduced, the odds of encountering big treasure type monsters are fairly low, the bigger monsters are weighted to be less common on the random encounter tables. Since treasure type is already connected to monster type (e.g., more powerful monsters have better odds of having magic items), this keeps you from rolling big items as often. So already there is a layer between “low level parties” and powerful treasure. The problem is, I suspect, that people look at this in isolation, at the treasure tables in the DMG, but not at the treasure type tables in the Monster Manual.

If there is ONE thing you CANNOT do with 1e AD&D, it is look at anything in isolation. 

Testing the Hypothesis

Let’s take a closer look at those treasure tables. So say you are concerned about a powerful item coming up for a low level party, what are the odds?

Let’s take a low level monster, to see the likelihood that a beatable low level opponent might have “big treasure”. Let’s assume a party of first level characters, say 6, encounter a giant spider, and the DM rolls that there is only 1. They find it in-lair. and after losing one party member to a poison bite, there are 5 who survive. 

A party of 6 first level PCs fighting a giant spider with one fatality and winning is a plausible AD&D scenario.

Treasure type for a giant spider is C:

20% - 1-12 1000s copper

30% - 1-6 1000’s silver

10% - 1-4 1000’s electrum

25% - 1-6 gems

20% - 1-3 jewelry

10% - Any 2 magic items

So the odds of getting nothing at all (e.g., failing to roll a successful result for any of these categories) are about 27%, so around ¾ of the time you will get something, but only 10% of the time will you get a magic item. You will most likely end up with SP or gems.

But let’s for the sake of argument assume that it all comes together, the party beats the spider, the party does so in lair so they find the loot, and the loot contains 2 items of magic. This is the “nightmare scenario” of the party pulling a powerful item from an encounter like this.

However, we first need to look at the treasure tables, “any two” means you roll twice on the table:

Magic Item Table A - DMG

01 -20 - Potions (A)

21 -35 - Scrolls (B.)

36-40 - Rings (C.)

41 -45 - Rods, Staves & Wands (D.)

46-48 - Miscellaneous Magic5 (E.l.)

49-51 - Miscellaneous Magic (E.2.) 

52-54 - Miscellaneous magic (E.3.) 

55-57 - Miscellaneous Magic (E.4.) 

58-60 - Miscellaneous Magic (E.5.)

61 -75 - Armor & Shields (F.) 

76-86 - Swords (G.) 

87-00 - Miscellaneous Weapons (H.) 

So what are the odds of a powerful item coming up? It varies by table, but note that the biggest odds here are for potions, scrolls are common as well, both by design. Scroll spells can be formidable, but 60% of the scroll table is spell scrolls, and for any given scroll with spells three of the four spellcasting classes won’t be able to use those spells. So if there is a party illusionist or druid the odds of getting scroll spells are very low.

It is possible to roll up a formidable item from any of the ‘big’ tables, e.g. rings, rods, staves and wands, misc magic, armor, weapons and swords. However, it is also possible to roll up a +1 sword, or a ring of protection +1, indeed these items are weighted more heavily than the others on the table to make them more likely to come up. 

The easiest way to show this would be to look at an individual table and determine rough odds of a “powerful” item coming up.

So say we rolled on the rings table, there is a roughly 40% chance of getting a “weaker” ring, e.g. a ring of warmth or a ring of protection +1. There is a 20% chance of getting a cursed ring. That leaves a 40% chance of getting a powerful ring.

So what are the odds of that happening? Say the party was traveling through a dense forest, that’s six checks of 1 in 10 odds of an encounter. What are the odds of encountering a giant spider and rolling up a powerful ring on the treasure table?

Odds of rolling a random encounter over the day - 50% 

Odds of rolling a giant spider random encounter - 5%

Odds of finding the spider in lair - 70%

Odds of rolling magic items - 10%

Odds of rolling rings - 5%

Odds of rolling powerful ring - 40%

Chance of rolling up a powerful ring after slaying a giant spider? About 0.0035%, since you get two rolls, that’s 0.0070%. 

That’s not really a fair way to look at it though, as you aren’t just rolling on the ring table. To expand this a bit, let’s assume that all of the tables are balanced so the weight of results is towards less powerful items, e.g., when you get one of the permanent item tables like the ring table, the odds of rolling one of the more powerful items is around 40% or less than half. A “powerful” item has some effect beyond a +1, e.g. +2 or higher, or a power of some kind (like a Horn of Valhalla). 

So the question becomes, what are the odds that a 1st level party traveling through the forest in a tropical environment will encounter a monster (as monsters have magic item loot), and then roll magic items? I will assume a 50% “in lair” as monster “in lair” odds vary wildly. I will also average out the percentage odds of getting a magic item as the odds vary wildly by treasure type, but in almost all cases (except for treasure type U,V,W and X) odds are below 50%. So to be generous I will average these to 50%.

Odds of rolling a random encounter over the day - 50% 

Odds of rolling a monster random encounter - 50%

Odds of finding the monster in lair - 50%

Odds of rolling magic items - 50%

Odds of rolling potentially powerful magic items (e.g. not potions)- 80%

Odds of rolling powerful item - 40%

Overall odds of a 1st level party ending up with a powerful item - 2%

So about 2 times in 100, your first level party will roll up a powerful magic item after a random encounter. And that’s not even considering the odds of them being defeated by the monster. A 4HD giant spider in its lair with a poison bite is nothing to sneeze at. 

Observations

Part of the point of reviewing odds, even when you have to make a lot of assumptions to calculate them, is to make it clear that the system as a whole makes powerful magic item acquisition from random encounters fairly rare. There are ample restrictions, animals don’t have magic items and make up a large chunk of the tables, the likelihood of magic items is calibrated to the power of the monster, magic items are often found in lair but the monsters aren’t there, even if they do have loot, the odds of magic items are very low, and the odds of powerful magic items are lower than less powerful items and cursed items.

More powerful monsters will be your best bet for finding more powerful items, but even then, you can roll weaker items or cursed items or items that can’t be used by any of the party members (depending on party composition). And of course there is the risk!

NPCs are a good source of magic items, but using the Appendix P system to generate their items means that they will skew towards temporary magic and low “+” weapons and armor.

So even though some magic items are rolled randomly at the table, and others are rolled randomly before the game, all are rolled randomly using these tables, and the odds of getting a powerful item at lower levels are very, very low. 

With all of that in mind, I have data from three years of using this system to assign magic items. ALL MAGIC ITEMS IN ALL OF MY CAMPAIGNS have been rolled randomly using the system I have outlined here using the weighted tables created by Gygax.

I thought I would give some samples of the magic item inventory from my different groups.

My Tuesday group PCs are 4-5th level. 

The party priest is 4th level and has the following magic items after a year of gaming:

+1 ring of protection

+1 hammer

Potion of ESP

Scroll: Silence 15’ rad., 

The party magic-user is 5th level and has the following magic items after a year of gaming:

+1 dagger

Scroll: Wizard Eye, Scroll: Monster Summoning III, Polymorph Other, Bigby’s Forceful Hand

Potion of Flying

My Wednesday group has PCs that vary from 5-6th. 

The 5th level ranger has the following magic items after 2 years of play.

Eyes of the Eagle

+2 sword

Potion of Healing

Potion of Levitation

The 6th level magic-user has the following items after 2 years of play:

Scroll: Fly, Invisibility 10’ rad.

Scroll: Contact Other Plane, Fear

Bracers of Defense AC:7

Ring of Protection +1

Dagger +1/+3 vrs large creatures

Wand of Frost (7 charges)

Potion of Gaseous Form

This PC has the most magic items of any PC in any of my games, and he is not the highest level PC (the highest level PC in any of my campaigns is 8th level), reinforcing the point that the number of magic items possessed by a PC is not directly tagged to level, part of the randomization piece. 

My Friday group has the highest level PCs of any of my groups as they have played the longest, the oldest PC in the group is an 8th level fighter after 3 years of play, his magic items are:

Chameleon Cloak

+2 axe

+1 sword +3 vrs giant creatures

Potion of Sweet Water

Potion of Healing

The party priest in the Friday group is 8th level, has been playing for 3 years, and has the following items:

Staff of Curing (11 charges)

+1 mace

+1 chain mail

Scroll: Flame Strike, Insect Plague, Curse

Potion of Fire Resistance

A few clarifications. First, these are representative examples, but due to the nature of one-shot magic items like scrolls and potions, things can vary a lot. So for example, if one of the PCs picked up a scroll and happened to roll a scroll of 6 spells, they could temporarily have a significant boost in power. Also, magic items get destroyed, fireballs, pick pockets, falling into rivers, crushing blows, etc., etc., etc. So although there may have been more magic items than are listed here, they didn’t all last that long.

By the Numbers

The point of discussing this in fair detail was to show that rolling randomly for magic items, whether ahead of time for intelligent monsters/NPCs or on the spot for random encounters, does not “break” the game or generate any wild discrepancies in power. Instead, if you use the tables as they were designed, and rather than picking, roll randomly on them, they are weighted in such a way as to produce a very modest spread of treasure.

I haven’t talked about gold/loot here, because that’s another post, but suffice it to say if you use the tables and generate loot, and loot is the primary source of leveling in AD&D, you will get level progression as mentioned above:  

1 year of gaming (80hrs) - party members from 4-5th lv 

2 years - 5-6th

3 years 6-8th

Three years of gaming (240 hours of gaming, or 120 sessions of 2 hours each) to get a fighter to 8th level with one miscellaneous magic item, two low bonus magic weapons and two potions seems fairly reasonable to me. I’m not sure if people would consider this “low magic” or not, but I would definitely not call it “high magic”. 

What matters to me most about this system is that it randomizes treasure generation, so I don’t have to pick anything, but it is weighted randomization, so it doesn’t create wild results. I like this as it keeps things unpredictable, it saves me the time and effort of trying to determine what treasures “best fit” with the campaign, it produces a wider spread of items, and it is exciting to roll for some treasures live at the table.  

And it produces results where magic items are not TOO common, and PCs don’t end up walking around with dozens of magic items on their person. I think it’s a tribute to the genius of Gygax’s game design that the system works this well without me having to do the heavy lifting of curating the results for my game. 

It also reinforces my belief that the reason you get “Monty Haul” campaigns using AD&D is that referees are ignoring some aspect of this system. I have noted that the NPCs in many of Gygax’s modules have treasure spreads similar to those seen here. Bhakashal uses the AD&D system, the only addition to the system is that 50% of the time I roll on the Bhakashal tables instead of the AD&D tables, but the weightings are exactly the same, e.g. you have the same odds of rolling each individual type of magic item (e.g. rings, potions, misc items), and roughly the same distribution of powerful versus weaker/cursed items as the AD&D tables do.